Utah HERO Project Phase 1 Findings

Sept. 15, 2020

Compiled by the Sorenson Impact Center at the David Eccles School of Business in partnership with Utah HERO Project leadership.

Through May and June 2020, a team of researchers collaborated with the Utah governor’s office to conduct what is, to-date, the largest and most thorough randomized COVID-19 testing project in the United States. To conduct the testing, 82 faculty and staff — all part of the Utah HERO project — surveyed and tested about 10K people in Utah’s major population centers for both active infections and antibodies.

Results from the study, which aimed to determine the true prevalence of the virus, suggest about 40% of COVID-19 cases are detected by current testing measures, meaning there are about 1.5 undetected cases of COVID-19 in Utah for each detected case. 

Primary HERO project partners include the Marriner S. Eccles Institute for Economics and Quantitative Analysis at the David Eccles School of Business, University of Utah Health, ARUP Laboratories, Hope Corps, and the Utah Governor’s Office of Management and Budget.

Main findings

Results from random, representative COVID-19 testing across four Utah counties suggest about 40% of COVID-19 cases are detected by current testing measures. While many Utah residents experiencing symptoms consistent with Covid-19 will seek out clinical testing from the healthcare system to confirm their diagnosis. not all residents seek testing, and some carriers do not have active symptoms. 

The HERO Project’s community-wide testing and analysis found that for every clinical case detection, there were approximately 1.5 cases that were not detected. This ratio of undetected to detected cases is lower than reported in other community seroprevalence studies, meaning Utah’s testing performance in the early months of the pandemic Utah was more effective compared to other states. The 95% confidence interval for the ratio of undetected to detected cases is 0.0 to 4.0.

Additional findings: 

• 0.81% of Utahns age 12 or older—or about 1 in 124—in the four-county area have COVID-19 antibodies. It is important to interpret the results of all surveys, including the HERO Project, within the context of a margin of error which expresses the uncertainty in each result. This project expresses the margin of error in terms of 95% confidence intervals, which are defined to have a 95% chance of including the true result. The 95% confidence interval for seroprevalence in the four-county area is 0.15% to 1.61%.
• Infections of some vulnerable groups are three times higher. Across all project areas and subgroups in Phase 1, seroprevalence was 0.81%. Comparatively, seroprevalence of Hispanic Utahns was 2.73% — over three times higher. In Summit County, which reported a relatively early outbreak of cases, seroprevalence was 4.59%—over five times higher than the rest of the state. More accurate data about Covid-19 antibodies found among these populations helps Utah monitor equity gaps and develop more effective solutions to slow the spread for all communities in the state.
• About 0.3% of infections, detected or undetected, result in death. By estimating the number of undetected cases, HERO Project leaders were also able to provide a ballpark estimate of the infection fatality rate among all people infected with Covid-19, including both detected and undetected cases. The estimate of 1.5 undetected cases for each detected case translates to an infection fatality rate of about 0.3%, which gives Utah’s decision-makers more accurate information about potential fatality rates associated with future infections.

Why randomized testing is the gold standard approach for determining true prevalence

Randomized, representative testing can be costly and difficult to implement. But experts say it’s the only way to truly know how widespread the disease is in our communities. 

To start with, non-random testing only reveals the tip of the iceberg. Non-random testing — for example, testing available only to those with symptoms or exposures — often excludes individuals who choose not to seek care or face barriers to seeking care, those whose symptoms are too mild to be tested under current protocols, or those who may be asymptomatic but contagious. 

Second, representative testing ensures we have data on the full population, not just one demographic sub-set. Even so-called scientific studies that appear random because they use ‘convenience’ samples — for example, visitors to grocery stores or tests of blood samples for non-COVID-related tests — won’t be representative of the population. 

Third, randomized testing helps policymakers improve public health and economic relief measures. With gold-standard data in-hand, policymakers can initiate appropriate contact tracing, improve estimates of the risks of community-based transmission, and accelerate the treatment and self-isolation of cases at earlier stages of disease progression.